A FPGA Vein Imaging System Based on Matched Filter

Abstract

Venipuncture is one of the most frequently used medical means. Fat thickness and skin color of different people may increase the difficulty for medical staff to judge the position and depth of vein. Therefore, this paper proposes a near-infrared vein imaging system, which is an auxiliary device for venipuncture. Through acquiring the near-infrared images of skin, it can show the subcutaneous venous vessels and assist medical staff to improve the success rate of venipuncture. We use FPGA chip, which has great parallel computing ability, and noncoaxial optical structure of multiplexing to design the vein imaging system. In addition, we design a matched filter vessel enhancement algorithm based on FPGA, it can accurately retain the vessel depth information. Comparison with other filters verifies the effectiveness of this proposed algorithm. Experiments show that this vascular enhancement device has better image quality on the premise of ensuring the real-time and effective performance.